PIXE analysis of Zn enzymes

Nuclear Instruments and Methods in Physics Research B 150 (1999) 222±225 PIXE analysis of Zn enzymes C. Solõs a,* , A. Oliver a, E. Andrade a, J.L. Ruvalcaba-Sil a, I. Romero b, H. Celis c a b c Instituto de Fõsica, Universidad Nacional Aut onoma de M exico, Apdo. Postal 20-364, 01000 M exico, D. F., Mexico Facultad de Medicina, Departamento de Bioquõmica, Universidad Nacional Aut onoma de M exico, 04510 M exico, D. F., Mexico Instituto de Fisiologõa Celular, Universidad Nacional Aut onoma de M exico, Apdo. Postal 70-243, 01000 M exico, D. F., Mexico Abstract Zinc is a necessary component in the action and structural stability of many enzymes. Some of them are well characterized, but in others, Zn stoichiometry and its association is not known. PIXE has been proven to be a suitable technique for analyzing metallic proteins embedded in electrophoresis gels. In this study, PIXE has been used to investigate the Zn content of enzymes that are known to carry Zn atoms. These include the carbonic anhydrase, an enzyme well characterized by other methods and the cytoplasmic pyrophosphatase of Rhodospirillum rubrum that is known to require Zn to be stable but not how many metal ions are involved or how they are bound to the enzyme. Native proteins have been puri®ed by polyacrylamide gel electrophoresis and direct identi®cation and quanti®cation of Zn in the gel bands was performed with an external proton beam of 3.7 MeV energy. Ó 1999 Elsevier Science B.V. All rights reserved. Keywords: PIXE; Zinc; Enzymes; Electrophoresis 1. Introduction Metals are integral components of many enzymes. They play an important role in the enzymatic reactions of living organisms and as structural elements that give stability to the enzymes. One of the major problems to study the metal role in the protein is their quanti®cation and location because of the low amounts involved, usually under the detection limits of conventional analytical techniques. New physical methods with * Corresponding author: Tel: +525 6065159; fax: +525 6161535; e-mail: [email protected]®sicacu.unam.mx lower detection limits can help to overcome this problem. Proton induced X-ray emission (PIXE) has been very useful to identify the location of metals in proteins subunits [1±3] and to quantify them [4,5]. In this work we used PIXE to study Zn enzymes. Zinc is an essential component of more than 200 enzymes isolated from di€erent species. At present, the role of Zn in the catalytic activity of some of them is well understood, but less is known about the role of Zn as a structural element. For the present study, the metalloenzyme carbonic anhydrase (CA), of known Zn content, was chosen to establish the methodology and then we focused the study on the cytoplasmic (soluble) pyrophosphatase (PPase) from the photosynthetic 0168-583X/99/$ ± see front matter Ó 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 5 8 3 X ( 9 8 ) 0 1 0 1 2 - X C. Solõs et al. / Nucl. Instr. and Meth. in Phys. Res. B 150 (1999) 222±225 bacteria Rhodospirillum rubrum for which the number of Zn2‡ ions that bind to the enzyme is not known. Cytoplasmic pyrophosphatase is an enzyme found in all organisms from bacteria to man. This enzyme has a dual function. It breaks the high energy bond of pyrophosphate and replenishes the orthophosphate to the energy-converting systems [6]. R. rubrum pyrophosphatase requires free Zn for stabilization and its substrate is a metal ion± pyrophosphate complex. The number of Zn ions per protein was explored by PIXE. 2. Material and methods Carbonic anhydrase was obtained from Sigma Chemical Co. St Louis, MD. CA is a Zn metalloenzyme with a molecular weight of 29 000. It has one Zn coordinated with three histidines. Cells of the photosynthetic bacterium Rhodospirillum rubrum were grown anaerobically in light at 30°C in a medium described in Ref. [7]. Cytoplasmic pyrophosphatase (PPase) was extracted and partially puri®ed by a modi®cation of the procedure by Klemme and Gest [8]. It has a molecular weight of 120 000. Slab gels of 1.5 mm thickness were used. The stacking and resolving gels contained 4.5% and 7.5% acrylamide respectively. Proteins were resuspended under nondenaturing conditions in a bu€er containing 0.5 M Tris HCl pH 8.0, 64% glycerol and 0.04% blue bromophenol. PPase was incubated in a bu€er containing 20 mM Tris HCl pH 8.6 and 0.1 mM ZnSO4 before loading the gels and its activity was monitored by colorimetry as in Ref. [9]. Aliquots containing 15±300 lg of protein were loaded per gel slot. One portion of the gel was removed and stained to locate the proteins with coomassie brilliant blue, another was used to measure activity of the PPase and the rest of the gel was dried without staining for PIXE analysis. The gels were dried on a layer of cellophane and covered with a layer of polyethylene ®lm that was removed to expose the gel directly to the proton beam. Digital images of the stained gels were used to obtain the densitograms by computer analysis. 223 PIXE analyses of intact gels were carried out with an external proton beam of 3.7 MeV of energy generated by a 3 MV pelletron accelerator of the Instituto de Fõsica, UNAM. A graphite collimator of rectangular section of 8 ´ 1 mm2 was used. Each gel was placed at 1.5 cm from the exit of the beam and the currents were kept under 4 nA to avoid gel damage. The gel was moved in steps of 1 or 1.5 mm. The emitted X-rays were detected by a LEGe detector placed at 65° respect to the beam. Mylar (21 lm) and aluminum (38 lm) ®lters were placed in front of the detector to reduce the high counts due to Ar X-rays. The PIXE spectra were analyzed using the AXIL program. The counts due to Ar were used as a relative measure of the charge reaching the gels. Therefore, Ka X-ray intensities of Zn were normalized to Ar K X-ray intensities. 3. Results and discussion In order to determine the amount of Zn detected by PIXE in each band, calibration was carried out by preparing gels with di€erent amounts of ZnSO4
7H2 O (between 0 and 300 lg/g) [5]. Fig. 1 shows the relationship obtained between the Ka X-ray intensities of Zn normalized to the Ar K X-ray intensities as a function of the Zn concentration. The calibration curve is linear for Zn concentrations as high as 300 lg/g. This allows Fig. 1. Zn Ka X-ray yield/Ar K X-ray yield vs. Zn concentration in gels. 224 C. Solõs et al. / Nucl. Instr. and Meth. in Phys. Res. B 150 (1999) 222±225 us to determine Zn in proteins whose concentrations varies within a wide range. When compared to conventional biochemical methods used in Zn determination [10], PIXE is much more sensitive and has lower detection limits. In this work, we analyzed CA by PIXE which is a Zn enzyme very well characterized and that is used as an internal control to detect Zn in proteins prepared in PAGE [11]. We chose this enzyme to validate the conclusions derived from the analysis of PPase whose Zn content is unknown. The PIXE scan along the gel obtained for CA is shown in Fig. 2. Each point represents the Zn content (lg/g of dried gel) of one irradiated section of the band. The total amount of Zn detected in a whole band was computed following the procedure described by Weber et al. [5]: The total amount of Zn in the bombarded band can be calculated from the area `S' (in lg mm/g), under the Gaussian ®tted to the experimental points. From the gel mass per mm `MAC ' of irradiated gel, the total number of Zn atoms in the band `NZn ' is given by Ref. [5]: NZn ˆ MAC SNAV ÿ6 10 ; AWZn Where NAV and AWZn are the AvogadroÕs number and the Zn atomic weight respectively. On the other hand, knowing the amount of protein Mp contained in a particular band, the number of protein molecules Np can be calculated by the relationship: Np ˆ NAV Mp ; MWp Where MWp is the protein molecular weight. The ratio `R' of Zn atoms per protein molecules can then be determined: Rˆ NZn : Np In native gels of CA, three bands are present. It has been reported that these bands correspond to charge isomers and not to oligomers [11]. The di€erence in the migration distance of the isomers is interpreted in terms of a di€erence on the aminoacid composition. The Zn distribution measured Fig. 2. Zn distribution obtained by PIXE scanning of the gel containing 128 lg of carbonic anhydrase. The arrows indicate the position of the bands. by PIXE for CA is shown in Fig. 2. It can be seen that Zn is only detected in the ®rst band (the position of the three protein bands is indicated by arrows in the Fig. 2) and not in the other two bands. This is a surprising result since it was expected that the three bands would contain Zn. It has been reported that CA has one Zn per enzyme. However, the Zn distribution from Fig. 2 shows that Zn is present only in the ®rst band with an stoichiometry of 1 : 1 (Table 1). This fact suggests that the di€erence in mobilityÕs of the CA charge isomers is due to the presence of Zn. This will be studied in more detail in a future work. The Zn distribution in a gel of cytoplasmic PPase of R. rubrum is shown in Fig. 3. This enzyme, partially puri®ed has four bands in native PAGE stained for proteins with Coomassie blue. However, when the gels are speci®cally stained for PPase activity, only one band appears (its position is indicated by an arrow in Fig. 3). As shown in Table 1 Values for the metal/enzyme ratio (R) Enzyme R experimental R expected Carbonic anhydrase Carbonic anhydrase Pyrophosphatase Pyrophosphatase 1.3 0.7 2.1 1.9 1 1 Unknown Unknown C. Solõs et al. / Nucl. Instr. and Meth. in Phys. Res. B 150 (1999) 222±225 225 periods of time. It is also easier to locate the region to be bombarded. Regarding our results, this work provides a new example of the applicability of PIXE in the study of Zn enzymes. This is the ®rst time that the Zn association and stoichiometry is reported for the cytoplasm PPase of R. rubrum. These are preliminary results and more experiments including PPases from di€erent sources and other Zn enzymes are contemplated. Acknowledgements Fig. 3. Zn distribution obtained by PIXE scanning of the gel containing 300 lg of cytoplasmic pyrophosphatase of Rhodospirillum rubrum. The arrow indicates the position of the Pyrophosphatase band. Fig. 3, this PPase band has a well de®ned peak of Zn whose stoichiometry is calculated to be two Zn per protein (Table 1). However, another band considered as contaminant has also a certain amount of Zn associated, as can be seen in the Zn distribution. This suggests that this contaminant band corresponds to a Zn protein. Alternately, it could be an inactive PPase form. As mentioned above, PPase is known to be stabilized by Zn [8]. Using PIXE we have demonstrated the association of Zn to the PPase and that this enzyme contains two atoms of Zn. 4. Conclusion PIXE measurements in air has been proven to be an adequate technique to analyze Zn bound to enzymes embedded in polyacrylamide gels. In these conditions, charge build-up problems found in measurements performed under vacuum are avoided and irradiation can be done for longer This work was partially supported by the CONACyT grants F036-E9109, G0010-E and 127262-E. The authors would like to express their gratitude to K. L opez, E. Perez-Zavala and J.C. Pineda for maintenance and operation of the accelerator, and Silvia Escobedo for sample preparation. References [1] Z. Szokefalvi-Nagy, I. Demeter, Cs. Bagynka, K.L. Kovacs, Nucl. Instr. Meth. B 22 (1987) 156. [2] Z. Szokefalvi-Nagy, Nucl. Instr. Meth. B 109/110 (1996) 234. [3] C. Solõs, A. Oliver, E. Andrade, Nucl. Instr. Meth. B 136/ 138 (1998) 928. [4] Z. Szokefalvi-Nagy, Cs. Bagynka, I. Demeter, K.L. Kovacs, L.H. Quynh, in: Schrauzer (ed.), Biological Trace element Research. Humana Press, Clifton, NJ, 1990, p. 93. [5] G. Weber, D. Strivay, C. Menendez, B. Schoefs, M. Bertrand, PIXE 6 (1996) 215. [6] R. Lathi, Microbiol. Rev. 4 (1983) 169. [7] G. Cohen-Bazire, W.R. Sistrom, R.Y. Stainer, J. Cell Comp. Physiol. 49 (1957) 25. [8] J.H. Klemme, H. Gest Eur, J. Biochem. 22 (1971) 528. [9] I. Romero, A. Gomez-Pliego, H. Celis, J. Gen. Microbiol. 137 (1991) 2611. [10] E. Hilario, I. Romero, H. Celis, J. Biochem. Biophys. Methods. 21 (1990) 197. [11] SIGMA technical bulletin No MKR-137.